With the development of the economy and the surge in car ownership, the sale of used cars has been welcomed by more and more people, and the information of the vehicle condition is the focus information of them. The f...With the development of the economy and the surge in car ownership, the sale of used cars has been welcomed by more and more people, and the information of the vehicle condition is the focus information of them. The frame number is a unique number used in the vehicle, and by identifying it can quickly find out the vehicle models and manufacturers. The traditional character recognition method has the problem of complex feature extraction, and the convolutional neural network has unique advantages in processing two-dimensional images. This paper analyzed the key techniques of convolutional neural networks compared with traditional neural networks, and proposed improved methods for key technologies, thus increasing the recognition of characters and applying them to the recognition of frame number characters.展开更多
DQ impedance-based method has been widely used to study the stability of three-phase converter systems.As the dq impedance model of each converter depends on its local dq reference frame,the dq impedance modeling of c...DQ impedance-based method has been widely used to study the stability of three-phase converter systems.As the dq impedance model of each converter depends on its local dq reference frame,the dq impedance modeling of complex converter networks gets complicated.Because the reference frames of different converters might not fully align,depending on the structure.Thus,in order to find an accurate impedance model of a complex network for stability analysis,converting the impedances of different converters into a common reference frame is required.This paper presents a comprehensive investigation on the transformation of dq impedances to a common reference frame in complex converter networks.Four different methods are introduced and analyzed in a systematic way.Moreover,a rigorous comparison among these approaches is carried out,where the method with the simplest transformation procedure is finally suggested for the modeling of complex converter networks.The performed analysis is verified by injecting two independent small-signal perturbations into the d and the q axis,and doing a point-by-point impedance measurement.展开更多
Target signal acquisition and detection based on sonar images is a challenging task due to the complex underwater environment.In order to solve the problem that some semantic information in sonar images is lost and mo...Target signal acquisition and detection based on sonar images is a challenging task due to the complex underwater environment.In order to solve the problem that some semantic information in sonar images is lost and model detection performance is degraded due to the complex imaging environment,we proposed a more effective and robust target detection framework based on deep learning,which can make full use of the acoustic shadow information in the forward-looking sonar images to assist underwater target detection.Firstly,the weighted box fusion method is adopted to generate a fusion box by weighted fusion of prediction boxes with high confidence,so as to obtain accurate acoustic shadow boxes.Further,the acoustic shadow box is cut down to get the feature map containing the acoustic shadow information,and then the acoustic shadow feature map and the target information feature map are adaptively fused to make full use of the acoustic shadow feature information.In addition,we introduce a threshold processing module to improve the attention of the model to important feature information.Through the underwater sonar dataset provided by Pengcheng Laboratory,the proposed method improved the average accuracy by 3.14%at the IoU threshold of 0.7,which is better than the current traditional target detection model.展开更多
为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了...为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了编码器对CTU进行四叉树递归遍历划分的编码时间。原始编码策略中粗模式决策通过基于残差经哈德曼变换的预测残差绝对值总和(Sum of Absolute Transformed Difference,SATD)的损失值来估计率失真优化过程中的率失真损失值来进行加速,但仍会耗费一定的编码时间。提出一种方法通过采样搜索的方式减少粗模式决策过程中计算的模式数,从35种模式降低到了18种模式,降低了粗模式决策过程中计算估计损失值的时间。由粗模式决策过程得到的较优的多个候选帧内模式来进行率失真优化,为了缩减粗模式决策需要计算的候选模式数,在候选模式列表中根据前后帧内预测角度模式的估计损失值的差距来筛选掉部分可能性较低的候选模式实现早停止决策,从而减少需要进行率失真优化的候选模式数量,进而减少率失真优化过程的计算时间。本文提出的算法在测试序列上平均实现78.15%的编码时间缩减,BD-PSNR为-0.168 d B,BD-RATE为3.49%。展开更多
文摘With the development of the economy and the surge in car ownership, the sale of used cars has been welcomed by more and more people, and the information of the vehicle condition is the focus information of them. The frame number is a unique number used in the vehicle, and by identifying it can quickly find out the vehicle models and manufacturers. The traditional character recognition method has the problem of complex feature extraction, and the convolutional neural network has unique advantages in processing two-dimensional images. This paper analyzed the key techniques of convolutional neural networks compared with traditional neural networks, and proposed improved methods for key technologies, thus increasing the recognition of characters and applying them to the recognition of frame number characters.
基金The support of the first and fourth authors is given by National Key R&D Program of China,2018YFB0905200.The support for the second and third authors is coming from BIRD171227/17 project of the University of Padova.
文摘DQ impedance-based method has been widely used to study the stability of three-phase converter systems.As the dq impedance model of each converter depends on its local dq reference frame,the dq impedance modeling of complex converter networks gets complicated.Because the reference frames of different converters might not fully align,depending on the structure.Thus,in order to find an accurate impedance model of a complex network for stability analysis,converting the impedances of different converters into a common reference frame is required.This paper presents a comprehensive investigation on the transformation of dq impedances to a common reference frame in complex converter networks.Four different methods are introduced and analyzed in a systematic way.Moreover,a rigorous comparison among these approaches is carried out,where the method with the simplest transformation procedure is finally suggested for the modeling of complex converter networks.The performed analysis is verified by injecting two independent small-signal perturbations into the d and the q axis,and doing a point-by-point impedance measurement.
基金This work is supported by National Natural Science Foundation of China(Grant:62272109).
文摘Target signal acquisition and detection based on sonar images is a challenging task due to the complex underwater environment.In order to solve the problem that some semantic information in sonar images is lost and model detection performance is degraded due to the complex imaging environment,we proposed a more effective and robust target detection framework based on deep learning,which can make full use of the acoustic shadow information in the forward-looking sonar images to assist underwater target detection.Firstly,the weighted box fusion method is adopted to generate a fusion box by weighted fusion of prediction boxes with high confidence,so as to obtain accurate acoustic shadow boxes.Further,the acoustic shadow box is cut down to get the feature map containing the acoustic shadow information,and then the acoustic shadow feature map and the target information feature map are adaptively fused to make full use of the acoustic shadow feature information.In addition,we introduce a threshold processing module to improve the attention of the model to important feature information.Through the underwater sonar dataset provided by Pengcheng Laboratory,the proposed method improved the average accuracy by 3.14%at the IoU threshold of 0.7,which is better than the current traditional target detection model.
文摘为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了编码器对CTU进行四叉树递归遍历划分的编码时间。原始编码策略中粗模式决策通过基于残差经哈德曼变换的预测残差绝对值总和(Sum of Absolute Transformed Difference,SATD)的损失值来估计率失真优化过程中的率失真损失值来进行加速,但仍会耗费一定的编码时间。提出一种方法通过采样搜索的方式减少粗模式决策过程中计算的模式数,从35种模式降低到了18种模式,降低了粗模式决策过程中计算估计损失值的时间。由粗模式决策过程得到的较优的多个候选帧内模式来进行率失真优化,为了缩减粗模式决策需要计算的候选模式数,在候选模式列表中根据前后帧内预测角度模式的估计损失值的差距来筛选掉部分可能性较低的候选模式实现早停止决策,从而减少需要进行率失真优化的候选模式数量,进而减少率失真优化过程的计算时间。本文提出的算法在测试序列上平均实现78.15%的编码时间缩减,BD-PSNR为-0.168 d B,BD-RATE为3.49%。
